The specific aims of this epidemiologic investigation are: (1) To quantitate the magnitude of the associations between dietary factors and site-specific cancer rates among women, with primary attention given to the relationships of amount and type of dietary fat with breast cancer of vitamin A intake with cancer of all sites, and of the consumption of fat, fiber and meat with colon cancer. (2) To examine the relationships between intake of variety of other nutrients (such as cholesterol and Vitamins C and E as well as of specific food items (such as coffee, artificial sweeteners) and rates of a variety of cancers. (3) To quantify possible causal associations between other potential risk factors (including use of oral contraceptives, other hormones, prescription medications and hair dyes) and cancers of the breast, ovary, uterus, lung and melanoma. Interactive effects will be considered explicitly. These aims will be achieved by continuing the prospective follow-up of a cohort of 121,964 women aged 30-55 in 1976. Biennial mailed questionnaires record exposure and disease histories, and cancer diagnoses are confirmed by review of pathology reports and specimens. In 1980, initial detailed diet histories were obtained which have been shown to be reproducible by repeated administration of the questionnaire to a sub-sample of the cohort. We propose to continue to gather, by questionnaire every two years, data on diet, other exposures (particularly use of post-menopausal hormones), and incident fatal and non-fatal cancers. To date, the overall follow-up rate exceeds 90% and over 95% of incident cancers will be documented. The large sample size and high follow-up rate of this prospective cohort study will allow precise and valid estimates of the effects of various dietary components and other risk factors on cancer causation or prevention. These results, combined with existing knowledge, should allow health authorities to make appropriate recommendations regarding optimal diet and health practices.